Anthony Minessale
|
afd0e1fc01
|
add t38_pass_broken_boolean to pass the brokeb boolean behaviour over to the other side of the call when detected
|
2011-10-12 15:28:08 -05:00 |
|
Anthony Minessale
|
9e09483589
|
add manual_rtp_bug gen_one_gen_all to prevent rtp passthru to break the *S* word..
|
2011-10-07 10:40:30 -05:00 |
|
Brian West
|
7f04f7ace9
|
check decode here
|
2011-09-28 00:35:06 -05:00 |
|
Anthony Minessale
|
639afde2a4
|
FS-3571 --resolve
|
2011-09-23 08:20:59 -05:00 |
|
Anthony Minessale
|
842ea87d64
|
FS-3571 try this, I don't have a way to test it so this is just guessing, give me a lab box if this doesn't work
|
2011-09-21 14:56:49 -05:00 |
|
Anthony Minessale
|
626b5cb2a2
|
FS-3571 FS-3442 FS-957 --resolve tighten up the parsing of these boolean vals issue in t38 sdp
|
2011-09-19 20:11:23 -05:00 |
|
Anthony Minessale
|
4048a087b7
|
more 2833 recover code
|
2011-09-16 15:35:16 -05:00 |
|
Anthony Minessale
|
4d1a76ee5c
|
save telephony event data for recovery
|
2011-09-16 14:30:32 -05:00 |
|
Anthony Minessale
|
dfbb22fe91
|
alternate fix to f479462026
|
2011-09-16 14:16:31 -05:00 |
|
Anthony Minessale
|
f479462026
|
use bridge_to instead of signal_bond for recovering bridge
|
2011-09-15 07:56:19 -05:00 |
|
Brian West
|
06eb4b6ab3
|
Fix recovery when nat is involved
|
2011-09-09 02:47:56 -05:00 |
|
Anthony Minessale
|
5154b8819d
|
fix wrong media ip in recover data issue
|
2011-09-08 17:13:28 -05:00 |
|
Anthony Minessale
|
7acddfac95
|
FS-3549 --resolve This patch will probably make it work but the bug is actually in the phone, the patch is simply tolerating the bad behaviour. You are correct about the a=sendonly missing, this was fixed in a later revision of the polycom firmware. I suggest that even if this patch works, that you update your phones to a newer firmware, preferably the most recent.
|
2011-09-08 10:24:05 -05:00 |
|
Anthony Minessale
|
74d5a593af
|
revert forgot jira info
|
2011-09-08 10:23:40 -05:00 |
|
Anthony Minessale
|
361b623937
|
This patch will probably make it work but the bug is actually in the phone, the patch is simply tolerating the bad behaviour. You are correct about the a=sendonly missing, this was fixed in a later revision of the polycom firmware. I suggest that even if this patch works, that you update your phones to a newer firmware, preferably the most recent.
|
2011-09-08 10:22:46 -05:00 |
|
Anthony Minessale
|
7d399cce96
|
don't turn X-FS- headers into variables, they are reserved for FS specific communication and should not be passed on
|
2011-09-08 10:02:16 -05:00 |
|
Anthony Minessale
|
83a78fbfa0
|
FS-3545 --resolve no, that was not the intention, I missed a few more spots to hack in the exception for iLBC, (thanks for marring my code iLBC ppl) it should work as expected now even with the m_per_ptime on
|
2011-09-08 09:48:21 -05:00 |
|
Anthony Minessale
|
aea22cd4b7
|
add RTP_BUG_ACCEPT_ANY_PACKETS to disable dropping invalid packets for interop with Oracle CCA
|
2011-09-08 08:57:28 -05:00 |
|
Anthony Minessale
|
247537a959
|
FS-3545 --resolve this is actually compliant when mixing ptimes in the same sdp but since iLBC uses its own fmtp for ptime I will add this patch to make it beleive its 20 for the sake of arguement. If you have any other problems with this, set the channel or global variable sdp_m_per_ptime=false to completely disable the default correct behaviour
|
2011-09-07 08:07:04 -05:00 |
|
Anthony Minessale
|
4f58ad4fb7
|
misplaced !
|
2011-09-01 12:25:08 -05:00 |
|
Anthony Minessale
|
bff473bf32
|
antibiotics for our latest sonus infection
|
2011-08-31 19:30:58 -05:00 |
|
Anthony Minessale
|
61ab84396b
|
add new channel cap flag to indicate you can support queueable dtmf delay aka parsing w and W and implement it in the RTP stack and sofia endpoint
|
2011-08-29 11:04:38 -05:00 |
|
Anthony Minessale
|
d83245c1bd
|
fix logic determining which ip to use in contact on re-invites in recovery situations or when remote ip is unknown
|
2011-08-26 14:10:23 -05:00 |
|
Anthony Minessale
|
cab02bbc70
|
fix logic determining which ip to use in contact on re-invites in recovery situations or when remote ip is unknown
|
2011-08-26 13:51:41 -05:00 |
|
Anthony Minessale
|
5a062bc142
|
FS-3523 --resolve
|
2011-08-25 13:29:51 -05:00 |
|
Anthony Minessale
|
e644e620c8
|
reflect video fmtp on 1 legged calls
|
2011-08-25 09:24:17 -05:00 |
|
Anthony Minessale
|
a0c63e8bdc
|
FS-3522 --resolve
|
2011-08-25 08:25:47 -05:00 |
|
Anthony Minessale
|
0e5b694c44
|
add PAUSE_BETWEEN_DTMF rtp bug for sonus to delay sending 2833 digits so their buffering is not overwhealmed
|
2011-08-24 15:36:59 -05:00 |
|
Anthony Minessale
|
d481bf5eff
|
FS-3514 --resolve
|
2011-08-23 12:38:22 -05:00 |
|
Anthony Minessale
|
8abc3ba607
|
switch to INFO instead of nothing when no 2833 was found in sdp
|
2011-08-23 10:51:14 -05:00 |
|
Anthony Minessale
|
7386662164
|
rework session timers to try to have us not be the refresher whenever possible
|
2011-08-22 10:59:33 -05:00 |
|
Anthony Minessale
|
7159f64b22
|
revert 53b0ecce60
|
2011-08-19 17:29:47 -05:00 |
|
Anthony Minessale
|
850f2e3cff
|
FS-3510 --resolve make all instances of execute_on_X uniform with api_on_X and support both prefix vars as before and usage of arrays to push several exexute_on or api_on records into one variable
|
2011-08-18 09:11:45 -05:00 |
|
Anthony Minessale
|
53b0ecce60
|
use the call_id of the original register in the unsolicited notify for MWI
|
2011-08-05 11:09:45 -05:00 |
|
Anthony Minessale
|
0a97462f3f
|
FS-3340 revert
|
2011-08-04 00:01:26 -05:00 |
|
Anthony Minessale
|
77413ba9ba
|
FS-3451 --resolve we support this but there was a bug, should work now
|
2011-08-01 20:08:04 -05:00 |
|
Anthony Minessale
|
9e399c19e3
|
FS-3439 --resolve
|
2011-08-01 16:29:37 -05:00 |
|
Anthony Minessale
|
56d67eadf6
|
sdp_m_per_ptime is now implied to be true, if you don't like this set it to false but its going to be undefined behaviour. This basically means if you call in with ptime 30 then you have a bunch of ptime 20 codecs in your outbound list that there will be one m= line with 30 and the original inbound codec and more m= lines for each discinct ptime in your list. This is, of course, will depend on disable_trancoding or absolute_codec_string as well
|
2011-08-01 10:22:55 -05:00 |
|
Anthony Minessale
|
98473085ea
|
add homer capture hooks to mod_sofia
|
2011-07-31 18:36:23 -05:00 |
|
Anthony Minessale
|
f1de377d01
|
reapply revert
|
2011-07-16 11:37:03 -05:00 |
|
Anthony Minessale
|
1729078f56
|
FS-3386
|
2011-07-16 11:30:36 -05:00 |
|
Anthony Minessale
|
2f950634ab
|
fix sip auto_simplify
|
2011-07-16 02:34:17 -05:00 |
|
Anthony Minessale
|
1675981428
|
FS-3386 roll back a few revs then remove some refs and reroll patches that were in between
|
2011-07-16 01:03:01 -05:00 |
|
Anthony Minessale
|
30ebe5dea4
|
FS-3386 this should do it
|
2011-07-15 00:23:07 -05:00 |
|
Anthony Minessale
|
f9969f3856
|
only list real profiles not aliases in presence code
|
2011-07-14 16:01:44 -05:00 |
|
Anthony Minessale
|
2932c1fa17
|
FS-3386 Try this revision please
|
2011-07-14 00:17:05 -05:00 |
|
Anthony Minessale
|
7cb095190a
|
FS-3362 check for ports that are 0 in sdp and do not bother patching it for these cases
|
2011-06-24 18:28:56 -05:00 |
|
Anthony Minessale
|
bc7cb400c0
|
add sip_liberal_dtmf chanvar and liberal-dtmf profile param to use the maximum methods of DTMF avoiding sticking to the spec which leads to incompatability
|
2011-06-23 17:57:10 -05:00 |
|
Anthony Minessale
|
fb68746eed
|
add parallelism to sofia by offsetting sip messages to the concerned sessions and using multiple queue threads for message handling
|
2011-06-16 14:37:22 -05:00 |
|
Anthony Minessale
|
9df8169d1f
|
add mutex around gateway access on per-profile basis and token based access to global profiles to prevent hanging on to the hash mutex while doing sql stmts which may cause issues/slowdowns
|
2011-06-16 14:32:14 -05:00 |